Responsibilities
- Own end-to-end architecture for enterprise-scale GenAI and AI-powered solutions within the Self-Service Platform.
- Architect agentic AI applications with LangChain, LangGraph, orchestration patterns, prompt strategies, guardrails, and structured outputs.
- Design and optimize RAG pipelines involving chunking, embeddings, retrieval, re-ranking, and enterprise knowledge grounding.
- Integrate and optimize foundation models through Azure OpenAI, AWS Bedrock, and on-premises LLMs with routing and fallback strategies.
- Define GenAI reference architectures and evaluate LLMs, embedding models, vector databases, and orchestration frameworks.
- Embed security, privacy, Responsible AI governance, PII handling, data access controls, and content guardrails into AI designs.
- Build scalable Python backend APIs using FastAPI and asyncio with resilient Redis and RabbitMQ integrations.
- Guide MLOps/LLMOps standards for deployment, monitoring, retraining, drift handling, testing, security scanning, and infrastructure as code.
- Define LLM evaluation strategies and implement observability and tracing with Arize and LangSmith.
- Design memory strategies with retention and replay safety for long-running assistants.
- Containerize and deploy services using Docker and Kubernetes and govern Terraform-based infrastructure.],
- requirements exclamation_fix?[],
- benefits_fix?[]
- tech_stack_fix?[]
- category_rationale_fix?
- category_fix?
- specialties_fix?[]
- seniority_rationale_fix?
- seniority_fix?[]
- credential_category_fix?
- months_of_experience_fix?
- experience_in_place_of_education_fix?
- salary_fix?
Requirements
- BE/B.Tech/MCA in Computer Science, Engineering, or an equivalent qualification is required.
- 15+ years of software architecture experience and 3+ years designing AI/ML or LLM-based systems in production are required.
- Deep expertise in Python, FastAPI, asyncio, distributed services, and cloud-native architectures is required.
- Experience with RAG pipelines, embeddings, and vector databases such as Elastic, Pinecone, Milvus, and Chroma is required.
- Hands-on experience with LangChain, LangGraph, LlamaIndex, AutoGen, and interoperability patterns such as Model Context Protocol is required.
- Knowledge of LLM architectures, fine-tuning techniques including LoRA and PEFT, Azure OpenAI, and/or AWS Bedrock is required.
- Strong understanding of MLOps/LLMOps, Responsible AI principles, and governance is required.
- Proficiency with Docker, Kubernetes, Terraform, and cloud-native deployment practices is required.
- Experience with PyTorch, TensorFlow, Arize, LangSmith, AKS, Key Vault, MemGPT, LangMem, knowledge graphs, or telecom AI adoption is preferred.
- Academic credentials must be from recognized and accredited institutions and are subject to verification.
Benefits
- Positions are available in Bangalore, Kolkata, Gurgaon, Noida, and Chennai, with the primary city listed as Noida, India.
- Ericsson promotes an inclusive workplace and is an Equal Opportunity Employer.
Tech Stack
Categories
About Ericsson
The future of mobile isn’t on the horizon, it’s happening now. At Ericsson, we’re building the foundation for an open network ecosystem where industries, developers, and enterprises thrive. The convergence of 5G, AI, cloud, and network APIs isn’t just a technological shift; it’s a transformation that is redefining industries and enhancing everyday life. Open, programmable networks are enabling real-time innovation and unlocking new business models across the globe. Imagine a world where developers can dynamically access network capabilities on demand, where enterprises don’t just use connectivity but shape it. This isn’t a distant vision, it’s the ecosystem we’re creating today. Collaboration fuels everything we do. By working across industries, we’re designing a future where connectivity isn’t just seamless. It’s intelligent, programmable, and transformative. The shift is happening. Are you part of it?